Overview

There two programs to consider in Illinois for LMI population estimation: Solar for All (SFA) and the Microsoft Engie. Each program has unique geo- and LMI qualifications.

Geoqualification

The Microsoft Engie project uses a geoqualification map that was designed by Solstice. More details on the data used to create this map can be found on the Coda Page.

For the SFA Program, an initial set of census tracts were designed (“legacy” layer in the online map), but an updated version was made available (“current” layer). Both layers can be used until June 2024. Each of these layers are shown on the below map.

Note in the below table, you’ll see the Engie map covers nearly 62% of the state’s population. The combined coverage of SFA using both the legacy and current layers will result in roughly 45% of the state being covered; however, the current SFA map uses census tracts from the 2020 census, whereas the legacy SFA map uses census tracts from the 2010 census. This makes

LMI Qualification

For the Engie program, we use 100% AMI, SNAP and Medicaid households to estimate LMI populations. For the SFA program, the requirements slightly differ with a lower AMI percentage of 80%, with the same SNAP and Medicaid use cases.

LIFT Solar

There are 13 community solar projects in New Jersey according to the LIFT solar database.

Parcel Identification

Using Cook County’s Commerical data database, we pull all 2022 values for completeness. Using the property rate codes from the Assessor’s Office, we arrive at total non-residential (i.e. commercial) parcels. Note this data is only available for Cook County.

We tag the commercial parcels to qualified census tracts by both Engie and SFA. 77.6% of commercial parcels in Cook County are qualified by Engie, whereas 64.2% are qualified by SFA (combined method). The map below shows a sample of 10,000 commerical parcels tagged by either Engie or SFA and their relative locaiton.

Olive Township Analysis

townships <- st_transform(townships, st_crs(add_geo))

intersected <- st_intersection(add_geo, townships)

## Warning: attribute variables are assumed to be spatially constant throughout
## all geometries

intersected$area_intersected <- st_area(intersected)

add_geo$total_area_tract <- st_area(add_geo)

temp <- add_geo %>%
 # mutate(area = st_area()) %>%
  st_drop_geometry()

intersected <- intersected %>%
  st_drop_geometry()

temp <- left_join(temp, intersected[, c("total_pop", "AMI_80", "area_intersected")], by = c("total_pop", "AMI_80"))

temp$area_ratio <- temp$area_intersected / temp$total_area_tract

temp$total_pop_adjusted <- temp$total_pop * temp$area_ratio
temp$AMI_80_adjusted <- temp$AMI_80 * temp$area_ratio


final_temp <- temp %>%
  mutate(temp = 1) %>%
  group_by(temp) %>%
  summarize(total_pop_adjusted=sum(total_pop_adjusted, na.rm=T),
            AMI_80_adjusted = sum(AMI_80_adjusted, na.rm=T))
final_temp

## # A tibble: 1 × 3
##    temp total_pop_adjusted AMI_80_adjusted
##   <dbl>                [1]             [1]
## 1     1             14104.           1991.

Madison County

mad_geos <- add_geo %>%
  filter(grepl("Madison County", name))